MERLIN uses key moments to route multi-state linear attention
Reported long-context QA gains are +8.3 and +7.0 points on Gated DeltaNet-1.3B and Mamba-2-780M, with larger LongBench lifts.
2 stories, newest first · one card per story
Reported long-context QA gains are +8.3 and +7.0 points on Gated DeltaNet-1.3B and Mamba-2-780M, with larger LongBench lifts.
Training-free reasoning-plus-retrieval framework lifts multilingual multimodal entity linking accuracy by 6.9% overall and up to 23.3% on rare entities.
Today 1.6K readers · 2.4K views · 775 reads · all time 189K views